Continuation of analyses of the long term course of severe mental illness is proposed. Analyses to date have focussed on long-term course observed in four psychiatric case registers, with follow-up periods between 7 1/2- 20 years. An unusual date set from Madras, India, consists of ten years of monthly clinical ratings on 18 items, and yearly Present State Examinations (PSE's) for 90 schizophrenics who had their first onset of schizophrenia during 1981-1982. PSE's were administered by Dr. R. Thara, who was trained in its use by Indian collaborators on the International Pilot Study of Schizophrenia. Sixty-six of the original 90 will have complete monthly data sets for ten years. It is projected that 11 of the 85 now alive, but with incomplete data, will participate in the ten year follow- up, after intensive efforts at location and recruitment. The expected follow-up rate is 77 out of 85 still alive, (over 90%). Specific aims are: (1) Using varying definitions of episode, estimating and drawing Kaplan-Meier curves for lengths of episodes and remissions; (2) Cross-sectional and longitudinal syndromes will be sought using new methods of latent structure analysis; (3) Varying definitions of outcome will be predicted using socio-demographic indicators, signs and symptoms from early episodes, and syndrome definitions which result from Aim 2. These analyses have the potential for a deeper understanding of the natural course of schizophrenia than is available through any other data set. Generalization beyond the setting of India is feasible, due to the availability of standardized diagnostic and symptom data; but must be undertaken with some caution. The results will provide new insight into the heterogeneity of schizophrenia. Findings will be useful to clinicians in treating patients with seemingly unpredictable course.